Typically developing infants have fundamental perceptual biases that direct their attention to socially relevant stimuli such as voices and faces Children diagnosed with Autism Spectrum Disorder (ASD) show apparent deficits in auditory and visual processing of the voices and faces of their own species (conspecifics). We propose a prospective, experimental, longitudinal approach examining whether infants who lack foundational perceptual biases to attend to auditory and visual characteristics of conspecifics are more likely to develop deficits in social communication skills that are characteristic of ASD symptomology. By focusing on foundational and species-typical perceptual processes rather than complex skills, we hope to uncover earlier reliable markers and a mechanistic understanding compared with existing studies with high-risk siblings. This proposal has two short-term goals. The first goal is to help identify very young infants who have an increased risk for later social impairments typical of ASD by examining how they perceive voices and faces, the socially-relevant characteristics of their own species. The second goal is to uncover mechanisms by which deficits in fundamental biases might increase risk for later social impairments typical of ASD. This prospective experimental approach dovetails with current observational approaches such as parent- and physician- reported checklists, but by examining developmental mechanisms could direct intervention. To meet these goals, we focus on later-born siblings of children diagnosed with ASD (who have a 19% risk of recurrence, and increased probability of cognitive or language difficulties), and a comparison group of typically developing siblings. We will test whether infants show a listening preference for biological stimuli at 4- 6 months (human speech over non-speech sounds, human faces over non-faces); whether they show a preference for conspecific stimuli at 7-9 months (human speech over non-human primate vocalizations, human faces over non-human primate faces); and if they understand functional aspects of speech and faces at 10-12 months (matching speech to human faces, and understanding that speech allows people to communicate). Across these ages, we will collect observational measures of parental input, global cognitive development, linguistic development, and ASD behaviors using the Autism Observation Scale- Infants, and follow children until potential diagnosis at 24 and 36 months using the Autism Diagnostic Observation Schedule. We will examine group differences between infant groups. We will examine individual infants' developmental trajectories, linking early biases to later abilities, an examine the mediating role of parental input. This project's long-term goal is to help early identification of atypical development at an individual level in high-risk infants before overt ASD behaviors emerge by focusing on fundamental perceptual biases, and ultimately guide early intervention across a broad ASD phenotype. This research thus aims to identify high-risk infants before the full ASD behavioral profile can be characterized through standard observational protocols.